Based on phase and level errors translating and interpolating correction to improve the accuracy of wind power forecasting

At present, the accuracy of short-term wind power forecasting is low, level errors are commonly used to evaluate index, while few statistics about phase errors(delayed or ahead of time)are studied. In this paper, a method based on translating and interpolating correction of phase and level errors was put forward to improve forecasting accuracy, and proposed the conception of phase error constant trend duration, and classified the day-ahead wind power forecasting curve and the actual curve into three modes: the constant trend of power increasing, decreasing and unchanging, the calculation model of constant trend duration was established, that based on probabilistic and statistical methods to obtain deviated value and lagged or lead direction between the forecasting and the actual time value. Using the translated and interpolated method, combining phase and level error to correct the day-ahead power forecasting error. The simulated result showed : the wind power forecasting phase error was 3.78h, and the direction was delay, and level absolute error was 40.05MW, what the phase error value that after the translated and interpolated correction was 2.56h, the error of which was reduced by 32.34%, level absolute error was 32.58MW, reduced by 18.65%, which effectively improved the accuracy of short-term wind power forecasting.